Abstract
This study examines how much the Child and Adolescent Mental Health Disorders Classification System of diagnosis overlaps with diagnoses in the Clinical Classification Software and the ICD-9-CM and ICD-10-CM.
Effective October 2015, all US health care institutions and practitioners transitioned to the International Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM) coding system, raising concerns about the validity of examining trends over time in clinical care, costs, and quality.1 For child mental health disorders, alignment with psychiatric diagnosis groups in the Diagnostic and Statistical Manual of Mental Disorders (Fifth Edition) (DSM-5) is also required to consistently examine trends. This is because the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) and ICD-10-CM diagnosis codes differ for individual disorders, and child mental health disorders are reclassified under new psychiatric diagnosis groups. We thus developed the Child and Adolescent Mental Health Disorders Classification System (CAMHD-CS), which classifies child mental health disorders across coding systems and aligns with DSM-5 psychiatric diagnosis groups. To examine our system’s performance, we compared detection of these disorders in a Medicaid claims database against the Clinical Classification Software (CCS) groups2 and then examined identification of the disorders across ICD-9-CM and ICD-10-CM coding systems using a national pediatric hospitalization database.
Methods
Following the general equivalency mapping guidelines,3 all ICD-9-CM and ICD-10-CM mental health diagnoses, source code descriptions, and corresponding DSM-5 categories were flagged as equal, approximate, or no match, and codes in the ICD-9-CM that linked to more than 1 ICD-10-CM code were identified. Backward and forward reference mappings were iteratively reviewed by a team of 2 board-certified child psychiatrists (B.T.Z. and C.R.), a senior pediatric researcher (J.C.G.), a statistician (J.R.), and a coding expert (A.D.). Of the 19 DSM-5 diagnostic groups, 15 were maintained, 3 were combined to create sexuality and gender identity disorders, and the former category neurodevelopmental disorders was broken out to include 7 discrete child-onset diagnosis groups. The ICD-10-CM codes that specified intentional self-harm were used to create the suicide or self-injury group. The remaining ICD-10-CM codes were grouped into 6 problem categories: namely, accidental or undetermined poisoning; maternal mental illness or substance abuse during pregnancy, delivery, or post partum; substance abuse–associated medical illness; fetal or newborn damage associated with maternal substance abuse; child mental health symptoms (eg, irritability); and miscellaneous conditions (eg, academic underachievement). The complete set of codes is available online.4
To compare the results of the child mental health disorder groups with an existing adult-based classification scheme, the CCS level 1 mental illness category and child mental health disorder groups were applied to the 2016 IBM Medicaid MarketScan. The sample included children age 3 to 17 years with mental health coverage and at least 11 months of enrollment who had a diagnosis of a CCS or child mental health disorder group on any inpatient or emergency department encounter or 2 outpatient encounters.5 To examine the performance of the CAMHD-CS across ICD-9-CM and ICD-10-CM, we compared the proportions of primary child mental health disorders in 2012 and 2016 hospitalizations among children age 3 to 17 years using the Kids’ Inpatient Database, a nationally representative sample of pediatric hospitalizations.2 Data were analyzed between May 2019 to June 2019 with SAS version 9.4 (SAS Institute).
Results
Among the 11 CCS groups, 21 child mental health disorder groups were specified (Table 1). The proportions of disorders identified across classifications were similar after accounting for CCS groups that have multiple disorder groups (Table 1). The proportions of primary child mental health disorders identified were also similar across ICD-9-CM and ICD-10-CM among US pediatric hospitalizations, with the exceptions of primary depressive disorder (ICD-9-CM, 58 038 individuals [36.5%]; ICD-10-CM, 91 241 individuals [48.4%]) and suicide or self-injury (ICD-9-CM, 251 individuals [0.2%]; ICD-10-CM, 17 617 individuals [9.3%]) (Table 2).
Table 1. Proportion of 4 124 483 Children Ages 3 and 17 Years in 2016 IBM Medicaid MarketScan Identified as Having an Encounter for a Primary Child Mental Health Disorder by Clinical Classification Software and Child and Adolescent Mental Health Disorders Classification System Groups.
| Clinical Classification Software | Child and Adolescent Mental Health Disorders Classification System | ||
|---|---|---|---|
| Primary child mental health disorder | Patients, No. (%) | Primary child mental health disorder | Patients, No. (%) |
| Any diagnostic groupa | 780 036 (18.9) | Any diagnostic groupa | 777 053 (18.8) |
| Attention-deficit, conduct, disruptive behavior | 420 928 (10.2) | Attention-deficit/hyperactivity disorder | 360 459 (8.7) |
| Disorders in infancy, childhood, or adolescence | 80 805 (2.0) | Autism spectrum disorder | 54 668 (1.3) |
| Motor disorders | 21 749 (0.5) | ||
| Developmental disorders | 215 954 (5.2) | Communication disorders | 111 194 (2.7) |
| Developmental delayb | 43 139 (1.0) | ||
| Intellectual disability | 27 079 (0.7) | ||
| Specific learning disorders | 98 674 (2.4) | ||
| Schizophrenia/other psychotic disorders | 7202 (0.2) | Schizophrenia spectrum and other psychotic disorders | 6922 (0.2) |
| Mood disorders | 133 217 (3.2) | Depressive disorders | 108 948 (2.6) |
| Bipolar and related disorders | 23 195 (0.6) | ||
| Anxiety disorders | 132 246 (3.2) | Anxiety disorders | 94 257 (2.3) |
| Obsessive-compulsive and related disorders | 4770 (0.1) | ||
| Adjustment disorders | 112 626 (2.7) | Trauma and stressor-related disorders | 150 283 (3.6) |
| Dissociative disorders | 106 (0.0) | ||
| Somatic symptom and related disorders | 1588 (0.0) | ||
| Screening/history mental health/substance abuse | 6985 (0.2) | Feeding and eating disorders | 3748 (0.1) |
| Sexuality and gender identity disorders | 912 (0.0) | ||
| Sleep-wake disorders | 5387 (0.1) | ||
| Substance-related disorders | 24 635 (0.6) | Substance-related and addictive disorders | 25 623 (0.6) |
| Delirium, dementia, amnestic/other cognitive | 3770 (0.1) | Neurocognitive disorders | 8057 (0.2) |
| Personality disorders | 3828 (0.1) | Personality disorders | 3414 (0.1) |
Includes problem categories (ie, suicide/self-injury, maternal mental illness, mental health symptom).
Includes unspecified neurodevelopmental disorder.
Table 2. Proportion of Pediatric Hospitalizations for a Primary Child Mental Health Diagnosis Across ICD-9-CM (2012) and ICD-10-CM (2016) Coding Systems Using Child and Adolescent Mental Health Disorders Classification System.
| Primary child mental health disorder | Kids’ Inpatient Database, No. (%) | |
|---|---|---|
| ICD-9-CM (n = 1 258 799) | ICD-10-CM (n = 1 092 787) | |
| Any diagnostic groupa | 159 057 (12.6) | 188 549 (17.3) |
| Attention-deficit/hyperactivity disorder | 5388 (3.4) | 5447 (2.9) |
| Autism spectrum disorder | 1939 (1.2) | 2682 (1.4) |
| Communication disorders | 40 (0) | 53 (0.0) |
| Developmental delayb | 55 (0) | 40 (0.0) |
| Intellectual disability | NA | 45 (0.0) |
| Motor disorders | 400 (0.3) | 319 (0.2) |
| Specific learning disorders | 27 (0.0) | 23 (0.0) |
| Schizophrenia spectrum and other psychotic disorders | 6721 (4.2) | 5198 (2.8) |
| Bipolar and related disorders | 17 897 (11.3) | 15 475 (8.2) |
| Depressive disorders | 58 038 (36.5) | 91 241 (48.4) |
| Suicide or self-injuryc | 251 (0.2) | 17 617 (9.3) |
| Anxiety disorders | 2807 (1.8) | 3367 (1.8) |
| Obsessive-compulsive and related disorders | 347 (0.2) | 387 (0.2) |
| Trauma and stressor-related disorders | 9896 (6.2) | 11 190 (5.9) |
| Dissociative disorders | 37 (0.0) | 22 (0.0) |
| Somatic symptom and related disorders | 1672 (1.1) | 1132 (0.6) |
| Feeding and eating disorders | 2576 (1.6) | 3221 (1.7) |
| Elimination disorders | 284 (0.2) | 237 (0.1) |
| Sleep-wake disorders | 130 (0.1) | 54 (0.0) |
| Sexuality and gender identity disorders | 71 (0.0) | 90 (0.0) |
| Disruptive, impulse control, and conduct disorders | 11 597 (7.3) | 11 948 (6.3) |
| Substance-related and addictive disorders | 3617 (2.3) | 2513 (1.3) |
| Neurocognitive disorders | 870 (0.6) | 560 (0.3) |
| Personality disorders | 294 (0.2) | 260 (0.1) |
Abbreviations: ICD-9-CM, International Classification of Diseases, Ninth Revision, Clinical Modification; ICD-10-CM, International Classification of Diseases, Tenth Revision, Clinical Modification; NA, not applicable.
Includes problem categories (ie, maternal mental illness, mental health symptom).
Includes unspecified neurodevelopmental disorder.
The ICD-9-CM codes identifying self-harm are E-codes and cannot appear as a primary diagnosis, whereas the ICD-10-CM specifies intentional self-harm by poisoning/overdose of drugs and medication.
Discussion
The CAMHD-CS builds capacity to identify child-onset mental health disorders for research using administrative data, disorders that would otherwise be underreported using CCS groups or the DSM-5 neurodevelopment disorder category. Limitations in our approach include using a modified general equivalency matching approach to accommodate examining corresponding DSM-5 diagnosis groups. Although the large proportion of pediatric hospitalizations for depression is consistent with prior studies,6 future research is required to examine the clinical validity, year-to-year trends, and contextual factors that influence the coding of diagnostic groups.
References
- 1.Reeves SL, Freed GL. Problems with quality measurement using International Statistical Classification of Diseases, Tenth Revision, Clinical Modification: the elephant no one know is in the room. JAMA Pediatr. 2019;173(6):515-516. doi: 10.1001/jamapediatrics.2019.0844 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Agency for Healthcare Research and Quality, Healthcare Cost and Utilization Project Beta clinical classifications software (CCS) for ICD-10-CM/PCS. Accessed June 13, 2019. https://www.hcup-us.ahrq.gov/toolssoftware/ccs10/ccs10.jsp
- 3.Center for Medicare & Medicaid Services 2014. ICD-10_CM and GEMs. Published 2014. Accessed August 9, 2019. https://www.cms.gov/Medicare/Coding/ICD10/2014-ICD-10-CM-and-GEMs.html
- 4.Children’s Hospital Association Mental health disorder codes. Published July 2, 2019. Accessed February 18, 2020. https://www.childrenshospitals.org/Research-and-Data/Pediatric-Data-and-Trends/2019/Mental-Health-Disorder-Codes
- 5.Witt WP, Fullerton CA, Chow C, et al. . Effect of having a usual source of care on health care outcomes among children with serious emotional disturbance. Acad Pediatr. 2017;17(1):45-52. doi: 10.1016/j.acap.2016.06.001 [DOI] [PubMed] [Google Scholar]
- 6.Bardach NS, Coker TR, Zima BT, et al. . Common and costly hospitalizations for pediatric mental health disorders. Pediatrics. 2014;133(4):602-609. doi: 10.1542/peds.2013-3165 [DOI] [PMC free article] [PubMed] [Google Scholar]